SEARCH RESULT

Data collection, data retention and analysis are becoming more and more important every day because of the fact that technology is involved almost all our life. The processing and analysis of the data can become more difficult with increasing precision. In three-dimensional surface modeling, due to the increase in sensitivity and data size depending on the surface state and extent of the surface, collecting and processing the data may become difficult. As a solution to this situation, we can see that Computational Geometry is used extensively. Computational Geometry derives intermediate interpolations by taking the start and end data as references instead of keeping each data separately. In this way, it is possible to model by determining intermediate values based on the mentioned reference points. There are various methods in Computational Geometry such as intersection detection, point position and triangulation. According to the needs, a solution way can be produced by various geometric computations. In our study, "Delaunay Triangulation" which is the most used type of triangulation methods will be examined.

International Conference on Advanced Technologies, Computer Engineering and Science
ICATCES

Mustafa Aksin Emrullah Demiral Ismail Rakıp Karas

One of the purposes of video surveillance systems is to detect anomalies which are unexpected situations at a certain location or at a frame. Anomalies can be related to motion or appearance according to its spatial position. In this paper, we propose an anomaly detection system based on spatio-temporal features. Features from Accelerated Segment Test (FAST) is used for detection of corners location. Optical Flow magnitude and orientation of these points is used as spatio-temporal features. A grid is to the frames to neutralize the effect of proximity to the camera. Normal patterns are clustered with an unsupervised neural network so called Self-Organizing Maps (SOM). In test videos if extracted features cannot model with normal clusters, associated grid cell will be marked as anomaly Keywords - Video Surveillance, Anomaly Detection, Features from Accelerated Segment Test (FAST), Optical Flow, SelfOrganizing Maps (SOM)

International Conference on Advanced Technologies, Computer Engineering and Science
ICATCES

Kadriye Öz Ismail Rakıp Karas

The importance of analysis high resolution satellite imagery plays an important research topic for geographical information analysis of cities. Geospatial data plays an important role in important issues such as governmental, industrial, research topics on traffic management, road monitoring, GNSS navigation, and map updating. In this study, road detection from satellite imagery methods are classified as classification-based, knowledgebased, mathematical morphology and dynamic programming. In the beginning, the road structures including feature and model are analyzed. Then, the advantages and disadvantages of road detection methods are evaluated and summarizez their accuracy and performance based on road detection principles. Therefore, in order to obtain remarkable results for road detection, it is better to use more than one method. In after days, performing a complex road extraction from a satellite image is still a necessary and important research topic.

International Conference on Advanced Technologies, Computer Engineering and Science
ICATCES

İdris Kahraman Ismail Rakıp Karas

Worldwide, vegetation cover functioning as the secure region for wild life, and natural water, air filter from pollution. Forecasting the vegetation dynamics assist the governments and managements to decrease the negative influence of vegetation dynamic fluctuations. In recent years, forecasting of precise vegetation dynamics become and highly important issue, due to rapid vegetation changings and the needs to protect this natural resource. The aim of this article is to forecasting the vegetation dynamics by applying neural networks (NN). Düzce region utilized as case study, which situated in the north west region of Turkey. Normalized difference vegetation index (NDVI) from Moderate Resolution Imaging Spectroradiometer (MODIS) were employed to create vegetation time series. From United States Geological Survey website, 300 NDVI interval data acquired and processed in ArcGIS software. The dataset of vegetation time series built based on required neural networks data structure. Spatiotemporal pixel based sampling strategy performed to forecast the vegetation dynamics. A number of geospatial data handling steps employed using Python and Matlab programing languages. Forecasting data separated to two subdivisions (training set, and testing set). Mean squared error (MSE) utilized as performance accuracy assessment metric. Neural networks effectively implemented using spatiotemporal data and achieve high testing accuracy. Consequences reveals the fitness of neural networks to forecast vegetation dynamics maps.

International Science and Engineering Application Symposium on Hazards
ISESH

S.K.M. ABUJAYYAB Ismail Rakıp Karas Emrullah Demiral

Interactive 3D architectural indoor design have been more popular after it benefited from Virtual Reality (VR) technologies. VR brings computer-generated 3D content to real life scale and enable users to observe immersive indoor environments so that users can directly modify it. This opportunity enables buyers to purchase a property off-the-plan cheaper through virtual models. Instead of showing property through 2D plan or renders, this visualized interior architecture of an on-sale unbuilt property is demonstrated beforehand so that the investors have an impression as if they were in the physical building. However, current applications either use highly resource consuming software, or are non-interactive, or requires specialist to create such environments. In this study, we have created a real-estate purposed low-cost high quality fully interactive VR application that provides a realistic interior architecture of the property by using free and lightweight software: Sweet Home 3D and Unity. A preliminary study showed that participants generally liked proposed real estate-purposed VR application, and it satisfied the expectation of the property buyers.

International Workshop on GeoInformation Science
GEOADVANCES

K. Ozacar Yasin Ortakcı I. Kahraman Rafet Durgut Ismail Rakıp Karas

Estimation and accurate assessment regarding population gets harder and harder day by day due to growth of world population in a fast manner. Estimating tendencies to settlements in cities and countries, socio-cultural development and population numbers is quite difficult. In addition to them, selection and analysis of parameters such as time, work-force and cost seems like another difficult issue. In this study, population number is guessed by evaluating light activities in İstanbul via night-time images of Turkey. By evaluating light activities between 2000 and 2010, average population per pixel is obtained. Hence, it is used to estimate population numbers in 2011, 2012 and 2013. Mean errors are concluded as 4.14% for 2011, 3.74% for 2012 and 3.04% for 2013 separately. As a result of developed thresholding method, mean error is concluded as 3.64% to estimate population number in İstanbul for next three years.

International Workshop on GeoInformation Science
GEOADVANCES

M. K. Turan E. Yücer E. Şehirli Ismail Rakıp Karas

Modern trendler arasında yer alan, internet dünyasında kullanıcının bilgiyi üretmesi ve tüketmesi coğrafi bilgi sistemleri içerisinde de yeni bir bakış açısı oluşturmaktadır. Coğrafi veri üretme işi günümüzde sadece uzmanlar ve profesyonellerin yapabileceği bir işlem değildir. Amatör internet kullanıcıları da hazırlanan çevrimiçi sistemler aracılığıyla coğrafi veriye katkı sağlamaktadırlar. Amatör kullanıcıların coğrafi veri üretmesi, gönüllü coğrafi bilgi olarak tanımlanır. Bu çalışmada gönüllü coğrafi bilgi sistemleri olarak kabul edilebilecek web sitelerinin, Foursquare gibi, verilerini görselleştirerek yayınlamış olduğu görseller üzerinden bir analiz yapılacaktır. Ayrıca temel olarak coğrafi bilgi sistemi olmayıp, yaptığı iş gereği bu veriyi oluşturan uygulamaların da, Uber gibi, ürettikleri veriler üzerinden yapılan görselleştirme de incelenecektir.

Akademik Bilişim
AB

Hacer Kübra SEVİNÇ Ismail Rakıp Karas

Son zamanlarda, yapay sinir ağları, diğer bütün alanlarda olduğu gibi eğitim alanında da önemli bir konuma gelmiştir. Gelişen teknoloji sayesinde eğitim kurumlarında öğrenci bilgi sistemlerinin kullanılmasıyla birlikte öğrencilerle ilgili her türlü veriye erişim sağlanmaktadır. Yapay sinir ağları, bu verileri kullanarak, diğer yöntemlerle mümkün olmayan analiz, hesaplama ve tahminleri başarılı bir şekilde yapabilmektedir. Bu çalışmada, Hitit Üniversitesi öğrenci bilgi sistemi veri tabanından 2014 - 2017 yılları arasındaki mezun olan öğrencilerin ortaöğretim, yükseköğretim, cinsiyet, yaş ve üniversite giriş sınav puanı verileri alınıp, Levenberg-Marquardt metodu kullanılarak yapay sinir ağı eğitilmiştir. Bu sayede öğrenci adayının, üniversitede seçmeyi düşündüğü bölüm ve programdaki göstereceği başarının (ne kadar sürede mezun olacağının ve mezun olduğunda elde edeceği genel not ortalamasının) yapay sinir ağı kullanılarak tahmin edilmesi sağlanmıştır.

Akademik Bilişim
AB

Hüseyin Çizmeci Ü. ATİLA Ismail Rakıp Karas